CN105765889A - Redundancy for port extender chains - Google Patents
Redundancy for port extender chains Download PDFInfo
- Publication number
- CN105765889A CN105765889A CN201580002350.4A CN201580002350A CN105765889A CN 105765889 A CN105765889 A CN 105765889A CN 201580002350 A CN201580002350 A CN 201580002350A CN 105765889 A CN105765889 A CN 105765889A
- Authority
- CN
- China
- Prior art keywords
- unit
- link
- port
- ring
- reserve link
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 239000004606 Fillers/Extenders Substances 0.000 title abstract 2
- 238000000034 method Methods 0.000 claims abstract description 20
- 230000004044 response Effects 0.000 claims description 8
- 238000005516 engineering process Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000015572 biosynthetic process Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 230000008672 reprogramming Effects 0.000 description 2
- 208000010378 Pulmonary Embolism Diseases 0.000 description 1
- 125000002015 acyclic group Chemical group 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000004899 motility Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000000284 resting effect Effects 0.000 description 1
- 238000011144 upstream manufacturing Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/38—Information transfer, e.g. on bus
- G06F13/40—Bus structure
- G06F13/4004—Coupling between buses
- G06F13/4027—Coupling between buses using bus bridges
- G06F13/4045—Coupling between buses using bus bridges where the bus bridge performs an extender function
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J3/00—Time-division multiplex systems
- H04J3/02—Details
- H04J3/08—Intermediate station arrangements, e.g. for branching, for tapping-off
- H04J3/085—Intermediate station arrangements, e.g. for branching, for tapping-off for ring networks, e.g. SDH/SONET rings, self-healing rings, meashed SDH/SONET networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/16—Handling requests for interconnection or transfer for access to memory bus
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F13/00—Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
- G06F13/14—Handling requests for interconnection or transfer
- G06F13/36—Handling requests for interconnection or transfer for access to common bus or bus system
- G06F13/362—Handling requests for interconnection or transfer for access to common bus or bus system with centralised access control
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L1/00—Arrangements for detecting or preventing errors in the information received
- H04L1/22—Arrangements for detecting or preventing errors in the information received using redundant apparatus to increase reliability
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/40—Bus networks
- H04L12/40006—Architecture of a communication node
- H04L12/40013—Details regarding a bus controller
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/42—Loop networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04J—MULTIPLEX COMMUNICATION
- H04J2203/00—Aspects of optical multiplex systems other than those covered by H04J14/05 and H04J14/07
- H04J2203/0001—Provisions for broadband connections in integrated services digital network using frames of the Optical Transport Network [OTN] or using synchronous transfer mode [STM], e.g. SONET, SDH
- H04J2203/0028—Local loop
- H04J2203/0039—Topology
- H04J2203/0042—Ring
Abstract
Techniques implementing redundancy in an extended bridge comprising a controller bridge (CB) unit and a plurality of port extender (PE) units are provided. In one embodiment, the CB unit can receive join requests from the plurality of PE units and can determine, based on the join requests, whether the plurality of PE units are physically connected to the CB unit and/or other CB units in the extended bridge according to a ring topology. If the plurality of PE units are physically connected to the CB unit or the other CB units according to a ring topology, the CB unit can select a link in the ring topology as being a standby link.
Description
The cross reference of related application
Subject application advocates that the title of application on October 31st, 2014 is the 62/073 of " redundancy (REDUNDANCYFORPORTEXTENDERCHAINS) of port expander chain ", the title of No. 832 U.S. Provisional Application cases and application on October 29th, 2015 is the priority of No. 14/926,822 U. S. application case of " redundancy (REDUNDANCYFORPORTEXTENDERCHAINS) of port expander chain ".The full content of this type of application case is to be incorporated herein by reference for all purposes.
Background technology
It is known that the group of the physical network devices that stack system or storehouse are (such as, via external cable) to link together for use as single logic network device in the field of computer networking.Each member network's device of storehouse is referred to as unit and comprises the stack-port of (1) other unit for being connected in identical storehouse, and (2) are for being connected to the FPDP of upstream/downstream main frame and/or network.
The stack system supporting IEEE802.1BR port expander technology (being called extension bridge) includes two kinds of unit: control bridge (CB) unit and port expander (PE) unit.CB unit provides for extension bridge and controls and management plane function.Such as, each CB unit can for having L2 switch or the L2/L3 router of appropriate managerial planar module (such as, CPU management etc.).CB unit can be connected to each other according to linear or ring topology via its stack-port, thus forming core stack.
Compared with CB unit, each PE unit is only extension bridge and provides physical data port terminal.Such as, each PE unit can be the L2 switch with X physical data port, and described physical data port shows as the virtual data port on the CB unit that PE unit is connected to.Because it is functional that PE unit need not perform any management plane, so this type of unit is generally less complicated, and therefore cost lower than CB unit.
Under existing 802.1BR standard, group's (being called PE chain) of PE unit can be connected to CB unit as linear sub-storehouse or tree.Such as, Fig. 1 describes to include the example extension bridge 100 of two CB unit CB1 and CB2.PE chain 102 is connected to CB1 as linear sub-storehouse, and wherein PE chain 102 comprises PE unit PE1, PE2, PE3, PE4 and PE5 (carrying out daisy chain connection from CB1 with described order).Each storehouse link between PE unit or between PE unit and CB unit includes uplink port (namely, it is provided that towards internuncial port of CB unit) and cascade port (namely, it is provided that along internuncial port of PE chain).Therefore, in the example of fig. 1, the port on the CB1 side of link 104 is cascade port, and the port on the PE1 side of link 104 is uplink port.Similarly, the port on the PE1 side of link 106 is cascade port, and the port on the PE2 side of link 106 is uplink port.The FPDP (that is, being connected to the port of other LAN assembly) of each PE unit is referred to as ECP Extended Capabilities Port.
One problem of the linear sub-stack arrangement shown in Fig. 1 is, if there is link or cell failure along PE chain 102, then some PE unit can be changed into inoperable.For example, it is assumed that there is link failure between PE unit PE2 and the PE3 extended in bridge 100.In this case, the connectivity that PE3, PE4 and PE5 will lose with CB1.If each PE unit comprises 48 FPDPs, then this will cause that 48x3=144 FPDP stops service, this may result in and destroys the network environment wherein disposing extension bridge significantly.
Above-mentioned link failure case can be included the junction line (being also called link aggregation group or LAG) of multiple physical port/link by formation between PE unit PE1 to PE5 and between CB unit CB1 and PE unit PE1 and be had a degree of alleviation.Due to this type of junction line in position in, even if a physical link breaks down, data still can continue to flow through remaining movable physical link along PE chain.But, relay and unresolved cell failure (that is, causing the fault that whole PE unit is closed).Such as, if PE unit PE2 breaks down or becomes inoperable, then the connectivity that PE3 to PE5 will lose with CB1, and unrelated with any relaying between described unit.
Summary of the invention
The invention provides the technology implementing to include the redundancy extended in bridge of CB unit and multiple PE unit.In one embodiment, described CB unit can receive from the plurality of PE unit and adds request and can ask to determine other CB unit whether the plurality of PE unit is physically connected to described CB unit and/or described extension bridge according to ring topology based on described addition.If the plurality of PE unit is physically connected to described CB unit or other CB unit described according to ring topology, then described CB unit may select the link in described ring topology as reserve link.
Described below and accompanying drawing provides the essence of specific embodiment and being best understood from of advantage.
Accompanying drawing explanation
Fig. 1 describes example extension bridge.
Fig. 2 describes the extension bridge with physics PE ring topology according to embodiment.
Fig. 3 A and 3B describes the workflow being used for disposing the addition of PE unit and generation logic PE chain according to embodiment.
Fig. 4 describes the workflow for disposing PE port/link fault according to embodiment.
Fig. 5 describes the network switch according to embodiment.
Detailed description of the invention
In the following description, for explanatory purposes, set forth that Several examples and details are to provide the understanding to each embodiment.But, it is appreciated by those skilled in the art that, some embodiment can be put into practice when without some these type of details, maybe can put into practice some embodiment with the amendment of some embodiment or equivalent.
1. general introduction
Embodiments of the invention provide the technology of the redundancy in the extension bridge of the chain for implementing to include being connected to the PE unit of the core stack of CB unit.At high-level place, this type of technology relates to being physically connected to PE chain core stack as ring (and nonlinearities storehouse), the one end making PE chain is connected to a CB unit in core stack, and the other end of PE chain is connected to another CB unit in core stack.Such as, Fig. 2 describes having CB unit CB1-A, CB2-S and CB3-M and including the extension bridge 200 of PE chain of PE unit PE17, PE18, PE19 and PE20 according to embodiment.CB1-A is the mobile C unit B in bridge 200, and CB2-S is standby CB unit and CB3-M is member's CB unit.As indicated, PE17 is physically connected to CB1-A and PE20 is physically connected to CB2-S, thus producing ring topology for PE chain.
When the existence of above-mentioned physical rings being detected, ring can be converted to one or more logic, linear PE chain by the CB unit (such as, mobile C unit B) of extension bridge.Such as, in fig. 2, the ring including PE unit PE17 to PE20 can be converted to the first logic linear chain comprising PE17 and PE18 and the second logic linear chain comprising PE19 and PE20 by mobile C unit B CB1-A.In order to realize this, CB1-A can link 202 (that is, connecting the link of two logic chains) between labelling/selection PE18 and PE19 as reserve link so that do not have internal bridge business to transmit along link 202.As used herein, " internal bridge business " refers to the business of header (such as, the ETAG header) labelling understood with only CB and PE unit.CB unit then can send a message to each PE unit to program its hardware list according to being sized logic PE chain, and the information about ring topology and logic PE chain can be stored in one or more local data structure (such as, topological database and ring table).In this way, extension bridge can be configured to carry out the operation of its uptime based on this logical topology.
At later time point, CB unit can to the one detection port in the unit in extension bridge, link or cell failure.In response to this fault, the physical rings structure of the available bridge of CB unit, and can automatically rearrange the logical topology of bridge to guarantee that each PE unit is still attached to core stack.Such as, return to above-mentioned case (wherein the link 202 of Fig. 2 is selected as reserve link), if connecting originating port or link failure subsequently on the link 204 of PE17 and PE18, then the reconfigurable logical topology of CB1-A makes PE17 form a PE chain (being connected to CB1-A) and PE18 to PE20 formation the 2nd PE chain (being connected to CB2-S).This can particularly relate to changes into activity (that is, operation) link by link 202 from reserve link, link 204 is changed into new reserve link, and the impacted PE unit (in this case, PE18 and PE19) of reprogramming is to reflect change.This of utilization logical topology reconfigures, and PE unit PE17 to PE20 can continue and at least one the CB unit communication in core stack, even if link 204 is turned off also being such.
So that lower part to describe this type of and the other side of the present invention in further detail.
2.PE unit adds and logic PE chain produces
Fig. 3 A and 3B describes the workflow 300 according to embodiment, and it can be performed by the CB unit (such as, the CB1-A of Fig. 2) extended in bridge when one or more PE unit attempts adding extension bridge.Workflow 300 assumes that PE unit is physically connected to the core stack of bridge as ring, such as the example of Fig. 2.Use workflow 300, CB unit can detect the existence of physical rings and ring can be converted to one or more logic, linear PE chain.
Starting with the frame 302 of Fig. 3 A, CB unit can add PE unit reception addition request (formatting according to (such as) Link Layer Discovery Protocol or LLDP) of extension bridge from hope.Add request and can comprise the link information (port that such as unit is connected by it) for PE unit and certification metadata.Add request and also can comprise the peculiar information of manufacturer, for instance the image version etc. that the types/models of unit, unit are just running.
Assume that each PE unit sends one from each non-data port (such as, cascade or uplink port) of device and adds request.Therefore, in the extension bridge 200 of Fig. 2, PE17 adds request by being connected in the port of CB1-A from it to send one, and from its be connected to the port of PE18 (towards CB2-S) send another and add request.If it is also supposed that PE unit receives from another PE unit adds request (such as, PE18 receives from PE17 and adds request), then described PE unit will forward addition request towards core stack when it has added extension bridge.If receiving PE unit not yet add extension bridge, then PE unit would first, through to send from its non-data port before forwarding forward is received addition request and joins request to attempt adding bridge.
At frame 304 place, in response to receiving addition request, CB unit can be set up with PE unit and control and status protocol (CSP) session and can through thus session and PE unit communication to carry out adition process according to 802.1BR standard.
At frame 306 place, CB unit can be used on the link information comprised in the addition request that frame 302 place receives and updates local topology database.Such as, CB unit can store about the information sending particular port and/or the link adding request above, and it indicates this PE unit is how to be physically connected to core stack.In this way, CB unit can follow the tracks of the overall physical topology of extension bridge.
Additionally, at frame 308 place, CB unit can use topological database to determine whether the interpolation of this new PE unit causes formation ring in physical topology.Without forming ring (frame 310), then CB unit can monitor additionally adding request (frame 312) and can loop back to frame 302 from other PE unit.If all PE of being connected unit have all added bridge and have been formed without ring, then workflow 300 can terminate.
But, if CB unit is determined at frame 310 place forms ring, then a link in the optional physical rings of CB unit is with acting on the reserve link that ring is separated into one or more logic, linear PE chain.Such as, in the example of figure 2, the link 202 between optional PE18 and the PE19 of CB1-A is as reserve link, thus ring is divided into the first logic linear chain including PE17 and PE18 and the second logic linear chain including PE19 and PE20.As mentioned previously, the link being selected as reserve link will not forward any inside (such as, the wrapping up ETAG's) bridge service in bridge.There is CB unit and can be used to select the various technology of reserve link (such as select to have added the last link of ring, select to be divided into ring the link etc. of evenly sized chain).One of ordinary skill in the art are it will be recognized that other changes, revises and substitutes.
Then, at frame 316 and 318 place, CB unit can update local ring table by the information (out of Memory together with such as selected reserve link) of the PE unit about makeup ring, and can send a message to each PE ring members with according to being sized logic linear chain execution hardware programming.In this way, each PE ring members can be configured to operate based on this logical topology.
In one embodiment, ring table can include an entry according to each PE ring members, has following field:
● PE unit ID
● HW program flag (true/false)
● ring members flag (true/false)
● ring ready flag (true/false)
● reserve link (port 1 arrives port 2)
HW program flag may indicate that the logic PE chain whether this PE unit has been determined according to frame 314 place programs within hardware.In general, once receive instruction hardware programming from PE unit to be successfully completed the response message of (below with respect to Fig. 4 discussion), this flag will be initialized to vacation and then be updated to true.Ring members flag may indicate that whether this PE unit is the part (will be initialized as very at frame 316 place) of physical rings.Ring ready flag may indicate that whether this ring of PE unit gets out when facing PE port/link/cell failure to support redundancy (that is, whether all link/unit in ring are whether in operation and through hardware programming).And the reserve link that recognizable frame 314 place of reserve link field determines.Because ring is ready and reserve link field will contain identical value generally for each PE ring members, it can be maintained in the ring table clause of single PE ring members (such as, having the PE unit of minimum ID) in certain embodiments.
It is transmitted into the HW of each PE ring members at frame 318 place by CB unit and programs message except about how, E-channel id (ECID) based on packet may also include, except forwarding the information of packet, the information indicating which kind of uplink port and cascade port to be applied to described unit (based on the logic PE chain that frame 314 is determined).Such as, about the extension bridge 200 of Fig. 2, it is assumed that CB1-A selects link 202 as reserve link, thus ring is divided into a logic linear chain with PE17 and PE18 and another logic linear chain with PE18 and PE20.It is further assumed that bridge 200 connects (the value x in attention port identifier x/y/z correspond to CB/PEID) with lower port:
● 1/1/1 to 17/2/1
● 17/2/3 to 18/2/3
● 18/2/1 to 19/2/1
● 19/2/3 to 20/2/3
● 20/2/1 to 2/1/1
In this case, following information can be programmed in the hardware of each PE unit:
●PE17
Zero uplink port: 17/2/1 (towards CB1-A)
Zero cascade port: 17/2/3 (towards PE18)
Zero for all ECID for PE17, forward port by for the corresponding data (that is, ECP Extended Capabilities Port) on PE17
Zero for all ECID for PE18, and forwarding port will be 17/2/3
Zero for all ECID for PE19, and forwarding port will be 17/2/3
Zero for all ECID for PE20, and forwarding port will be 17/2/3
●PE18
Zero uplink port: 18/2/3 (towards CB1-A)
Zero cascade port: nothing
Zero for all ECID for PE18, forward port by for the corresponding data (that is, ECP Extended Capabilities Port) on PE18
Zero for all ECID for PE17, and forwarding port will be 18/2/3
Zero for all ECID for PE19, and forwarding port will be 18/2/1
Zero for all ECID for PE20, and forwarding port will be 18/2/1
●PE19
Zero uplink port: 19/2/3 (towards CB2-S)
Zero cascade port: nothing
Zero for all ECID for PE19, forward port by for the corresponding data (that is, ECP Extended Capabilities Port) on PE19
Zero for all ECID for PE17, and forwarding port will be 19/2/1
Zero for all ECID for PE18, and forwarding port will be 19/2/1
Zero for all ECID for PE20, and forwarding port will be 19/2/3
●PE20
Zero uplink port: 20/2/1 (towards CB2-S)
Zero cascade port: 20/2/3 (towards PE19)
Zero for all ECID for PE20, forward port by for the corresponding data (that is, ECP Extended Capabilities Port) on PE20
Zero for all ECID for PE17, and forwarding port will be 20/2/3
Zero for all ECID for PE18, and forwarding port will be 20/2/3
Zero for all ECID for PE19, and forwarding port will be 20/2/3
Except above, CB1-A also can by as follows for other member (CB2-S and CB3-M) programming of himself and core stack:
● cascade port: 1/1/1 and 2/1/1
● the virtual port (VP) for PE17 and PE18 produces 1/1/1
● the VP for PE19 and PE20 produces 2/1/1
● notice that the VP of all generations all can reappear on all CB unit in core stack
With reference now to Fig. 3 B (continuing the workflow 300 from Fig. 3 A), send HW at frame 318 place and program message to after PE ring members, CB unit can receive response message from PE unit, and described response message indicates the programming on described unit to be successfully completed (frame 320).As response, the HW program flag of the entry of the described PE unit in ring table can be updated to very (frame 322) and can check whether all PE ring memberses have carried out HW programming (frame 324) by CB unit.If it is not, so workflow 300 can return to frame 320 to wait the response message from other ring members.
On the other hand, if at frame 324 place, CB unit determines that all PE ring memberses have carried out HW programming all, then the ring ready flag in ring table can be set to very (frame 326) by CB unit.This instruction hardware programming completes and PE ring members is ready for the offer redundancy when PE port/link/cell failure.Extension bridges and can start its normal forwarding operations and workflow 300 can terminate.
3. dispose PE port/link fault
Addition is extended bridge by the PE unit that the final result of workflow 300 is in physical ring topology, but a link in ring will be in " standby " pattern and therefore hinder internal bridge business (causing including the logical topology of two segregated linear PE chains).As mentioned previously, in one embodiment, the CB unit performing workflow 300 can only select that it is determined and will cause that annular becomes the first link of reserve link.In other embodiments, CB unit can select particular link as reserve link based on some factor (balanced load between active link in such as ring).Thereafter in a kind of situation, simple load balancing algorithm is in two chains being divided into by PE ring members and carrying approximately the same load.
When the PE port in bridge or any one in link break down subsequently, one or more CB unit (such as, the CB1-A of Fig. 2) in bridge can automatically rearranging logic topology and so that it is guaranteed that each operation PE unit be still attached to core stack.Fig. 4 describes the flow chart 400 for carrying out this process according to embodiment.
At frame 402 place, CB unit can receive the notice of PE port shutdown event.Non-working port can be physical port or logic port (such as, LAG).In the later case, port shutdown just can produce when notifying the generally only all physics port shutdowns in LAG.
At frame 404 place, CB unit can search for its ring table to determine the state of the ring belonging to PE unit with the PEID of the PE unit affected by port close event.If ring ready flag is set to false (instruction is not provided that redundancy) (frame 408), then CB unit can perform the nonredundancy of fault and dispose (frame 408) and workflow 400 can terminate.
If determining that ring ready flag is true at frame 408 place, then CB unit can check that whether the PE link affected by port close event is the reserve link (frame 410) of ring further.If it is then CB unit can only update its topological database to indicate port/link-down and to be set to false as (frame 418) by ring ready flag.Note this case need not perform any further step, because impacted/faulty link has been in resting state and has used thus without in logical topology.
But, if impacted/faulty link is current and the reserve link of acyclic, then the reconfigurable logical topology of CB unit to be labeled as reserve link and to be labeled as by old reserve link be movable (frame 412) now by impacted/faulty link.In this way, CB unit can produce two the new logics not affected, linear PE chain by port/link failure.
At frame 414 place, CB unit can send a message to impacted PE ring members new its corresponding hardware of logic PE chain reprogramming to produce according to frame 412 place/to determine, and can update the configuration of its local virtual port (VP) when needed.CB unit can also update its ring table so that impacted/faulty link to be identified as new reserve link (frame 416).
Finally, at frame 418 place, the renewable topological database of CB unit reflecting current connectivity in physical topology, the ring ready flag of ring is set to false as (because ring can not support any further redundancy when the second port/link fault occurs) and workflow 400 can terminate.
Each operation PE unit that the final result of workflow 400 is in ring still will be connected to core stack.Unique difference is that the logical topology of bridge will change to use impacted/faulty link as reserve link (may result in different logic PE chains).In general, this reconfigures and does not make any change in requisition for unit ID or ECID being assigned to PE unit；This type of ID can keep identical, and is how to be logically attached to core stack independent of PE unit.
It should be appreciated that Fig. 3 A, 3B and 4 workflow 300 and 400 be illustrative and without wishing to restriction embodiments of the invention.Such as, although the link that workflow 300 and 400 is assumed in the physical rings of PE unit is initially placed in standby mode, but does not operate link in alternative embodiments and become reserve link.It practice, the available data port on each PE unit is divided into two set, and the first set is by the first non-data link support of PE unit, and the second set is by the second non-data link support of PE unit.In this type of embodiment, if via the path fails of a link to core stack, then the port supported by described link can be changed to another link, thus providing the fullpath of core stack.
Although additionally, workflow 300 and 400 is described as being performed by the single CB unit (such as, activity/master unit) in core stack, but in alternative embodiments, this process can be disposed by multiple CB unit.In this type of embodiment, each CB unit can maintain the logic copy of topological database and ring table, and it synchronizes across core stack.
4. Instance failure case
This part list the extension various possible PE port/link fault case of bridge 200 about Fig. 2 and the configuration of the CB unit of bridge 200 and PE unit can be made so that dispose the change of this type of fault according to the workflow 400 of Fig. 4.This partly depends on the same port described in part 2 and connects, and assumes that link 202 (between 18/2/1 and 19/2/1) is initially selected as reserve link.
● the one in port 1/1/1 or 17/2/1 is closed
Zero PE ring is converted into the linear PE chain of the single logic including PE unit PE20, PE19, PE18, PE17
Zero port 17/2/1 is no longer set to uplink port
Zero port 17/2/3 becomes uplink port and 18/2/3 and becomes cascade port
Zero port 18/2/1 becomes uplink port and 19/2/1 and becomes cascade port
The VP of zero PE17 and PE18 moves to 2/1/1 from 1/1/1
Zero link 1/1/1 to 17/2/1 becomes new reserve link and ring ready flag is set to false
Zero topological database is updated over reflecting present physical topology
● the one in port 17/2/3 or 18/2/3 is closed
Zero PE ring is divided into two linear PE chains of logic: one is only be made up of PE17 and another one is made up of PE20, PE19, PE18
Zero port 18/2/1 becomes uplink port and 19/2/1 and becomes cascade port
The VP of zero PE18 moves to 2/1/1 from 1/1/1
Zero link 17/2/3 to 18/2/3 becomes new reserve link and ring ready flag is set to false
Zero topological database is updated over reflecting present physical topology
● the one in port 18/2/1 or 19/2/1 is closed
Zero link 18/2/1 to 19/2/1 remains reserve link and ring ready flag is set to false
Zero topological database is updated over reflecting present physical topology
● the one in port 19/2/3 or 20/2/3 is closed
Zero PE ring is divided into two linear PE chains of logic: one is only be made up of PE20 and another one is made up of PE17, PE18, PE19
Zero port 19/2/1 becomes uplink port and 18/2/1 and becomes cascade port
The VP of zero PE19 moves to 1/1/1 from 2/1/1
Zero link 19/2/3 to 20/2/3 becomes new reserve link and ring ready flag is set to false
Zero topological database is updated over reflecting present physical topology
● the one in port 20/2/1 or 2/1/1 is closed
Zero PE ring is converted into the linear PE chain of the single logic including PE unit PE17, PE18, PE19, EP20
Zero port 20/2/1 is no longer set to uplink port
Zero port 20/2/3 becomes uplink port and 19/2/3 and becomes cascade port
Zero port 19/2/1 becomes uplink port and 18/2/1 and becomes cascade port
The VP of zero PE19 and PE20 moves to 1/1/1 from 2/1/1
Zero link 20/2/1 to 2/1/1 becomes new reserve link and ring ready flag is set to false
Zero topological database is updated over reflecting present physical topology
5. the network switch
Fig. 5 describes the example network switch 500 according to embodiment.The network switch 500 can be used to implement any one in CB or the PE unit described in foregoing summary.
As indicated, the network switch 500 comprises management module 502, switch architecture module 504 and multiple I/O module 506 (1) to 506 (N).Management module 502 represents the control plane of the network switch 500 and therefore comprises one or more CPU management 508 of operation for managing/control device.Each CPU management 508 can be general processor, for instance based on PowerPC, Intel, AMD or ARM processor, under the control of its software in being stored in associated memory (not shown) operate.
Switch architecture module 504 and I/O module 506 (1) to 506 (N) jointly represent data or the Forwarding plane of the network switch 500.Switch architecture module 504 is configured to each other module of interference networks switch 500.Each I/O module 506 (1) to 506 (N) can comprise by the network switch 500 for sending and receive one or more input/output end port 510 (1) to 510 (N) of packet.As mentioned previously, port 510 (1) to 510 (N) can include the FPDP for communicating with other LAN assembly and for the stack-port with other unit communication in identical extension bridge.Each I/O module 506 (1) to 506 (N) also can comprise packet handler 512 (1) to 512 (N).Each packet handler 512 (1) to 512 (N) is hardware handles assembly (such as, FPGA or ASIC), and it can be made about how disposing the linear speed decision-making importing or deriving packet.
It should be appreciated that the network switch 500 is illustrative and without wishing to restriction embodiments of the invention.It is feasible for having more or less than other configurations many of the assembly of switch 500.
Above description illustrates each embodiment example together with the aspect that how can implement the present invention of the present invention.Above example and embodiment are not construed as simply embodiment, and are rendered as motility and advantage that the present invention as defined by the appended claims is described.Such as, although some embodiment describes about specific flow process and step, but is appreciated by those skilled in the art that, the scope of the present invention will not be limited strictly to described flow process and step.The step described in order can executed in parallel, the order of step can change and step can be modified, combines, adds or omit.As another example, although some embodiment is to use the incompatible description of particular group of hardware and software, it should be recognized that other combination of hardware and software is feasible, and the specific operation being described as implementing in software also can implement within hardware, and vice versa.
Description and graphic therefore should in illustrative implication but not restrictive sense is treated.Adopt other layout, embodiment, embodiment and equivalent are by the spirit and scope of the present invention understood for those skilled in the art and can set forth in without departing from such as appended claims.
Claims (19)
1. the method performed by the controller bridge CB unit in extension bridge, described method includes:
Addition request is received from multiple port expander PE unit by described CB unit；
Asked to determine other CB unit whether the plurality of PE unit is physically connected in described CB unit or described extension bridge according to ring topology based on described addition by described CB unit；And
If the plurality of PE unit is physically connected to described CB unit or other CB unit described according to ring topology, then select the link in described ring topology as reserve link.
2. method according to claim 1, wherein said reserve link will not forward the business within described extension bridge.
3. method according to claim 1, wherein said selection causes that the plurality of PE unit is divided into the linear PE chain of one or more logic.
4. method according to claim 2, it farther includes:
The ring table that described CB unit is local is updated to and comprises:
The instruction of described selected reserve link；
It is initialized to the first flag of vacation；And
For the entry of each in the plurality of PE unit, each entry includes the second flag being initialized to vacation.
5. method according to claim 4, it farther includes:
Message is transmitted into each in the plurality of PE unit, and described message comprises the instruction of the hardware for programming described PE unit according to one or more logic described linear PE chain.
6. method according to claim 5, wherein said message comprises the instruction for arranging the uplink port of described PE unit, cascade port and one or more forwarding port.
7. method according to claim 5, it farther includes:
Each from the plurality of PE unit receives instruction and is used for the response message that the described hardware of described PE unit has been successfully programmed；
Described first flag in described ring table is set to very；And
Described second flag in each entry of described ring table is set to very.
8. method according to claim 7, after it further includes at the port shutdown event about the PE unit in the plurality of PE unit that detects:
Whether described first flag checked in described ring table is set to true；And
If described first flag is set to true, then:
Determine by the link that described port shutdown event affects；And
Check whether described affected links is described reserve link.
9. method according to claim 8, if it farther includes described affected links is described reserve link, then:
Described first flag in described ring table is set to false as.
10. method according to claim 8, if it farther includes described affected links is not described reserve link, then:
Updating described ring table to indicate described affected links is described reserve link now, thus producing the linear PE chain of one or more new logic；
Described first flag in described ring table is set to false as；And
Sending a message to each in the plurality of PE unit, described message comprises the instruction of the described hardware for programming described PE unit according to one or more new logic described linear PE chain.
11. method according to claim 10, if it farther includes described affected links is not described reserve link, then:
The local virtual port VP configuration of described CB unit is updated according to one or more new logic described linear PE chain.
12. method according to claim 10, if it farther includes described affected links is not described reserve link, then:
Update the local topology database of described CB unit to indicate described affected links inoperable.
13. method according to claim 1, wherein said CB unit is the master unit in described extension bridge or active unit.
14. storage above has a non-transitory computer-readable storage medium for the program code that can be performed by the controller bridge CB unit in extension bridge, described program code causes described CB unit:
Receive from multiple port expander PE unit and add request；
Ask to determine other CB unit whether the plurality of PE unit is physically connected in described CB unit or described extension bridge according to ring topology based on described addition；And
If the plurality of PE unit is physically connected to described CB unit or other CB unit described according to ring topology, then select the link in described ring topology as reserve link.
15. non-transitory computer-readable storage medium according to claim 14, wherein said reserve link will not forward the business within described extension bridge.
16. non-transitory computer-readable storage medium according to claim 14, described link is wherein selected to cause that the plurality of PE unit is divided into the linear PE chain of one or more logic as reserve link.
17. an operable network equipment for use as the controller bridge CB unit in extension bridge, described network equipment includes:
Processor；And
Storage above has the memorizer of the program code that can be performed by described processor, and described program code causes described processor:
Receive from multiple port expander PE unit and add request；
Ask to determine other CB unit whether the plurality of PE unit is physically connected in described network equipment or described extension bridge according to ring topology based on described addition；And
If the plurality of PE unit is physically connected to described network equipment or other CB unit described according to ring topology, then select the link in described ring topology as reserve link.
18. network equipment according to claim 17, wherein said reserve link will not forward the business within described extension bridge.
19. network equipment according to claim 17, described link is wherein selected to cause that the plurality of PE unit is divided into the linear PE chain of one or more logic as reserve link.
Applications Claiming Priority (5)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| US201462073832P | 2014-10-31 | 2014-10-31 | |
| US62/073,832 | 2014-10-31 | ||
| US14/926,822 | 2015-10-29 | ||
| US14/926,822 US9984028B2 (en) | 2014-10-31 | 2015-10-29 | Redundancy for port extender chains |
| PCT/US2015/058258 WO2016070010A1 (en) | 2014-10-31 | 2015-10-30 | Redundancy for port extender chains |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105765889A true CN105765889A (en) | 2016-07-13 |
| CN105765889B CN105765889B (en) | 2018-05-29 |
Family
ID=55852814
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201580002350.4A Active CN105765889B (en) | 2014-10-31 | 2015-10-30 | Extension bridge and the method being executed by it |
Country Status (4)
| Country | Link |
|---|---|
| US (1) | US9984028B2 (en) |
| EP (1) | EP3213441B1 (en) |
| CN (1) | CN105765889B (en) |
| WO (1) | WO2016070010A1 (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107547410A (en) * | 2017-06-29 | 2018-01-05 | 新华三技术有限公司 | The method and device that a kind of multistage PE is cascaded automatically |
| CN108259296A (en) * | 2017-03-31 | 2018-07-06 | 新华三技术有限公司 | Message forwarding method and device |
| CN110741610A (en) * | 2017-01-09 | 2020-01-31 | 马维尔国际贸易有限公司 | Port expander with local switching |
| US11102115B2 (en) | 2017-05-31 | 2021-08-24 | New H3C Technologies Co., Ltd. | Forwarding packet |
| US11108594B2 (en) | 2017-05-31 | 2021-08-31 | New H3C Technologies Co., Ltd. | Implementing three-layer communication |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9792242B2 (en) * | 2014-12-09 | 2017-10-17 | Dell Products Lp | Systems and methods for non-unicast/destination lookup fail (DLF) load balancing |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102726006A (en) * | 2011-11-23 | 2012-10-10 | 华为技术有限公司 | Ethernet ring protection switching method, node and system |
| EP2670090A1 (en) * | 2012-05-31 | 2013-12-04 | Broadcom Corporation | Multi-homing in an extended bridge |
| CN103795518A (en) * | 2012-10-31 | 2014-05-14 | 杭州华三通信技术有限公司 | Method, device, and system for synchronizing port modes between devices |
| CN103856398A (en) * | 2012-11-30 | 2014-06-11 | 美国博通公司 | Virtual Trunking Over Physical Links |
| CN103873374A (en) * | 2014-03-27 | 2014-06-18 | 杭州华三通信技术有限公司 | Message processing method and device in virtualized system |
Family Cites Families (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6657951B1 (en) * | 1998-11-30 | 2003-12-02 | Cisco Technology, Inc. | Backup CRF VLAN |
| US6480473B1 (en) * | 1998-12-29 | 2002-11-12 | Koninklijke Philips Electronics N.V. | Verification of active nodes in an open network |
| US6760838B2 (en) * | 2001-01-31 | 2004-07-06 | Advanced Micro Devices, Inc. | System and method of initializing and determining a bootstrap processor [BSP] in a fabric of a distributed multiprocessor computing system |
| US7768904B2 (en) * | 2004-04-22 | 2010-08-03 | At&T Intellectual Property I, L.P. | Method and system for fail-safe renaming of logical circuit identifiers for rerouted logical circuits in a data network |
| ATE525829T1 (en) * | 2005-02-28 | 2011-10-15 | Ibm | BLADE SERVER SYSTEM HAVING AT LEAST ONE STACKED SWITCH WITH MULTIPLE SWITCHES CONNECTED TO EACH OTHER AND CONFIGURED FOR MANAGEMENT AND OPERATION AS A SINGLE VIRTUAL SWITCH |
| JP4760504B2 (en) * | 2006-04-12 | 2011-08-31 | 株式会社日立製作所 | Network system and communication device |
| JP2009188673A (en) * | 2008-02-05 | 2009-08-20 | Fujitsu Ltd | Transmitting apparatus and path setting method |
| JP2009205333A (en) * | 2008-02-27 | 2009-09-10 | Hitachi Ltd | Computer system, storage device, and data management method |
| CN101272352B (en) * | 2008-05-20 | 2012-01-04 | 杭州华三通信技术有限公司 | Looped network routing method and looped network node |
| US9032057B2 (en) * | 2009-04-06 | 2015-05-12 | Brocade Communications Systems, Inc. | Secure stacking setup implementing user interface of stacking device |
| CN101938395B (en) * | 2009-07-03 | 2014-08-13 | 中兴通讯股份有限公司 | Method and system for refreshing single ring address of Ethernet ring network |
| JP5521663B2 (en) * | 2010-03-15 | 2014-06-18 | 富士通株式会社 | COMMUNICATION DEVICE, COMMUNICATION SYSTEM, AND COMMUNICATION METHOD |
| CN102684979B (en) | 2011-03-11 | 2018-08-14 | 中兴通讯股份有限公司 | A kind of multicast data forwarding method and device for supporting virtual terminal |
| US20130100808A1 (en) * | 2011-10-24 | 2013-04-25 | Kaminario Technologies Ltd. | Managing Utilization Of A Logical Communication Path In A Multi-Path Channel |
| CN103297257B (en) * | 2012-02-27 | 2016-10-19 | 北京东土科技股份有限公司 | A kind of implementation method of redundant network |
| US20150372911A1 (en) * | 2013-01-31 | 2015-12-24 | Hitachi, Ltd. | Communication path management method |
| CN103220218B (en) * | 2013-04-28 | 2016-03-30 | 杭州华三通信技术有限公司 | The method and apparatus of anti-loops in vertical stack networking |
| CN104427012B (en) * | 2013-09-04 | 2018-12-11 | 新华三技术有限公司 | Port negotiation method and apparatus |
| CN104468854B (en) * | 2013-09-13 | 2017-10-27 | 新华三技术有限公司 | A kind of longitudinal fusion architecture VCF construction method and equipment |
| US9703747B2 (en) * | 2014-05-21 | 2017-07-11 | Dell Products Lp | Remote console access of port extenders using protocol extension |
-
2015
- 2015-10-29 US US14/926,822 patent/US9984028B2/en active Active
- 2015-10-30 WO PCT/US2015/058258 patent/WO2016070010A1/en active Application Filing
- 2015-10-30 EP EP15791882.2A patent/EP3213441B1/en active Active
- 2015-10-30 CN CN201580002350.4A patent/CN105765889B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102726006A (en) * | 2011-11-23 | 2012-10-10 | 华为技术有限公司 | Ethernet ring protection switching method, node and system |
| EP2670090A1 (en) * | 2012-05-31 | 2013-12-04 | Broadcom Corporation | Multi-homing in an extended bridge |
| CN103795518A (en) * | 2012-10-31 | 2014-05-14 | 杭州华三通信技术有限公司 | Method, device, and system for synchronizing port modes between devices |
| CN103856398A (en) * | 2012-11-30 | 2014-06-11 | 美国博通公司 | Virtual Trunking Over Physical Links |
| CN103873374A (en) * | 2014-03-27 | 2014-06-18 | 杭州华三通信技术有限公司 | Message processing method and device in virtualized system |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110741610A (en) * | 2017-01-09 | 2020-01-31 | 马维尔国际贸易有限公司 | Port expander with local switching |
| US11700202B2 (en) | 2017-01-09 | 2023-07-11 | Marvell Asia Pte Ltd | Port extender with local switching |
| CN108259296A (en) * | 2017-03-31 | 2018-07-06 | 新华三技术有限公司 | Message forwarding method and device |
| US11102115B2 (en) | 2017-05-31 | 2021-08-24 | New H3C Technologies Co., Ltd. | Forwarding packet |
| US11108594B2 (en) | 2017-05-31 | 2021-08-31 | New H3C Technologies Co., Ltd. | Implementing three-layer communication |
| CN107547410A (en) * | 2017-06-29 | 2018-01-05 | 新华三技术有限公司 | The method and device that a kind of multistage PE is cascaded automatically |
| CN107547410B (en) * | 2017-06-29 | 2020-05-12 | 新华三技术有限公司 | Method and device for automatic cascade of multi-stage PE |
Also Published As
| Publication number | Publication date |
|---|---|
| EP3213441B1 (en) | 2018-08-22 |
| CN105765889B (en) | 2018-05-29 |
| US9984028B2 (en) | 2018-05-29 |
| EP3213441A1 (en) | 2017-09-06 |
| US20160124884A1 (en) | 2016-05-05 |
| WO2016070010A1 (en) | 2016-05-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105765889A (en) | Redundancy for port extender chains | |
| CN101888333B (en) | Method and apparatus for maintaining port state tables in a forwarding plane of a network element | |
| WO2017092400A1 (en) | Failure recovery method and device, controller, and software defined network | |
| US8533316B2 (en) | Management computer for setting configuration information of node | |
| US20090201831A1 (en) | Vpls n-pe redundancy using pseudo wire fast failover | |
| CN104025513A (en) | Hierarchy of control in a data center network | |
| US11095476B2 (en) | Spanning tree protocol enabled n-node link aggregation system | |
| CN103188173B (en) | A kind of switch device | |
| CN101888307B (en) | The method and apparatus of port selection is realized for this locality | |
| CN102123104B (en) | Network device configuration correcting method and network device | |
| WO2017113929A1 (en) | Method and controller for controlling forwarding device cluster in software-defined networking | |
| CN105337780B (en) | A kind of server node configuration method and physical node | |
| CN105340230A (en) | Virtual chassis topology management | |
| CN110213107A (en) | A kind of autoplacement method, device and equipment for network topological diagram | |
| US7808888B2 (en) | Network fault correlation in multi-route configuration scenarios | |
| JP2008167315A (en) | Redundant line connecting method and wide-area communication network node device | |
| CN112165429B (en) | Link aggregation convergence method and device for distributed switching equipment | |
| CN104468831A (en) | Message service network cluster based on network topology | |
| CN107645402A (en) | A kind of route management method and device | |
| CN104394016A (en) | ISSU (In-Service Software Upgrade) method and system | |
| CN102291313B (en) | Method, device and equipment for finding virtual router redundancy protocol (VRRP) network topology | |
| US20190268235A1 (en) | Method for managing network nodes and communication control method thereof | |
| US8644137B2 (en) | Method and system for providing safe dynamic link redundancy in a data network | |
| CN102651712A (en) | Node routing method of multiprocessor system, controller and multiprocessor system | |
| CN102355370B (en) | Method for Ethernet board-level protection, device and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Address after: California, USA Applicant after: BROCADE COMMUNICATIONS SYSTEMS, Inc. Address before: California, USA Applicant before: BROCADE COMMUNICATIONS SYSTEMS, INC. |
|
| CB02 | Change of applicant information | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20180211 Address after: West Yorkshire Applicant after: Ares International IP Co.,Ltd. Address before: California, USA Applicant before: Brocade Communications Systems, Inc. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240109 Address after: London Patentee after: Andrew Wireless Systems UK Ltd. Address before: West Yorkshire Patentee before: Ares International IP Co.,Ltd. |
|
| TR01 | Transfer of patent right |